File size: 7,799 Bytes
45ee559
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
import os
import unittest

from tests import get_tests_input_path, get_tests_output_path, get_tests_path
from TTS.config import BaseAudioConfig
from TTS.utils.audio.processor import AudioProcessor

TESTS_PATH = get_tests_path()
OUT_PATH = os.path.join(get_tests_output_path(), "audio_tests")
WAV_FILE = os.path.join(get_tests_input_path(), "example_1.wav")

os.makedirs(OUT_PATH, exist_ok=True)
conf = BaseAudioConfig(mel_fmax=8000, pitch_fmax=640, pitch_fmin=1)


# pylint: disable=protected-access
class TestAudio(unittest.TestCase):
    def __init__(self, *args, **kwargs):
        super().__init__(*args, **kwargs)
        self.ap = AudioProcessor(**conf)

    def test_audio_synthesis(self):
        """1. load wav
        2. set normalization parameters
        3. extract mel-spec
        4. invert to wav and save the output
        """
        print(" > Sanity check for the process wav -> mel -> wav")

        def _test(max_norm, signal_norm, symmetric_norm, clip_norm):
            self.ap.max_norm = max_norm
            self.ap.signal_norm = signal_norm
            self.ap.symmetric_norm = symmetric_norm
            self.ap.clip_norm = clip_norm
            wav = self.ap.load_wav(WAV_FILE)
            mel = self.ap.melspectrogram(wav)
            wav_ = self.ap.inv_melspectrogram(mel)
            file_name = "/audio_test-melspec_max_norm_{}-signal_norm_{}-symmetric_{}-clip_norm_{}.wav".format(
                max_norm, signal_norm, symmetric_norm, clip_norm
            )
            print(" | > Creating wav file at : ", file_name)
            self.ap.save_wav(wav_, OUT_PATH + file_name)

        # maxnorm = 1.0
        _test(1.0, False, False, False)
        _test(1.0, True, False, False)
        _test(1.0, True, True, False)
        _test(1.0, True, False, True)
        _test(1.0, True, True, True)
        # maxnorm = 4.0
        _test(4.0, False, False, False)
        _test(4.0, True, False, False)
        _test(4.0, True, True, False)
        _test(4.0, True, False, True)
        _test(4.0, True, True, True)

    def test_normalize(self):
        """Check normalization and denormalization for range values and consistency"""
        print(" > Testing normalization and denormalization.")
        wav = self.ap.load_wav(WAV_FILE)
        wav = self.ap.sound_norm(wav)  # normalize audio to get abetter normalization range below.
        self.ap.signal_norm = False
        x = self.ap.melspectrogram(wav)
        x_old = x

        self.ap.signal_norm = True
        self.ap.symmetric_norm = False
        self.ap.clip_norm = False
        self.ap.max_norm = 4.0
        x_norm = self.ap.normalize(x)
        print(
            f" > MaxNorm: {self.ap.max_norm}, ClipNorm:{self.ap.clip_norm}, SymmetricNorm:{self.ap.symmetric_norm}, SignalNorm:{self.ap.signal_norm} Range-> {x_norm.max()} --  {x_norm.min()}"
        )
        assert (x_old - x).sum() == 0
        # check value range
        assert x_norm.max() <= self.ap.max_norm + 1, x_norm.max()
        assert x_norm.min() >= 0 - 1, x_norm.min()
        # check denorm.
        x_ = self.ap.denormalize(x_norm)
        assert (x - x_).sum() < 1e-3, (x - x_).mean()

        self.ap.signal_norm = True
        self.ap.symmetric_norm = False
        self.ap.clip_norm = True
        self.ap.max_norm = 4.0
        x_norm = self.ap.normalize(x)
        print(
            f" > MaxNorm: {self.ap.max_norm}, ClipNorm:{self.ap.clip_norm}, SymmetricNorm:{self.ap.symmetric_norm}, SignalNorm:{self.ap.signal_norm} Range-> {x_norm.max()} --  {x_norm.min()}"
        )

        assert (x_old - x).sum() == 0
        # check value range
        assert x_norm.max() <= self.ap.max_norm, x_norm.max()
        assert x_norm.min() >= 0, x_norm.min()
        # check denorm.
        x_ = self.ap.denormalize(x_norm)
        assert (x - x_).sum() < 1e-3, (x - x_).mean()

        self.ap.signal_norm = True
        self.ap.symmetric_norm = True
        self.ap.clip_norm = False
        self.ap.max_norm = 4.0
        x_norm = self.ap.normalize(x)
        print(
            f" > MaxNorm: {self.ap.max_norm}, ClipNorm:{self.ap.clip_norm}, SymmetricNorm:{self.ap.symmetric_norm}, SignalNorm:{self.ap.signal_norm} Range-> {x_norm.max()} --  {x_norm.min()}"
        )

        assert (x_old - x).sum() == 0
        # check value range
        assert x_norm.max() <= self.ap.max_norm + 1, x_norm.max()
        assert x_norm.min() >= -self.ap.max_norm - 2, x_norm.min()  # pylint: disable=invalid-unary-operand-type
        assert x_norm.min() <= 0, x_norm.min()
        # check denorm.
        x_ = self.ap.denormalize(x_norm)
        assert (x - x_).sum() < 1e-3, (x - x_).mean()

        self.ap.signal_norm = True
        self.ap.symmetric_norm = True
        self.ap.clip_norm = True
        self.ap.max_norm = 4.0
        x_norm = self.ap.normalize(x)
        print(
            f" > MaxNorm: {self.ap.max_norm}, ClipNorm:{self.ap.clip_norm}, SymmetricNorm:{self.ap.symmetric_norm}, SignalNorm:{self.ap.signal_norm} Range-> {x_norm.max()} --  {x_norm.min()}"
        )

        assert (x_old - x).sum() == 0
        # check value range
        assert x_norm.max() <= self.ap.max_norm, x_norm.max()
        assert x_norm.min() >= -self.ap.max_norm, x_norm.min()  # pylint: disable=invalid-unary-operand-type
        assert x_norm.min() <= 0, x_norm.min()
        # check denorm.
        x_ = self.ap.denormalize(x_norm)
        assert (x - x_).sum() < 1e-3, (x - x_).mean()

        self.ap.signal_norm = True
        self.ap.symmetric_norm = False
        self.ap.max_norm = 1.0
        x_norm = self.ap.normalize(x)
        print(
            f" > MaxNorm: {self.ap.max_norm}, ClipNorm:{self.ap.clip_norm}, SymmetricNorm:{self.ap.symmetric_norm}, SignalNorm:{self.ap.signal_norm} Range-> {x_norm.max()} --  {x_norm.min()}"
        )

        assert (x_old - x).sum() == 0
        assert x_norm.max() <= self.ap.max_norm, x_norm.max()
        assert x_norm.min() >= 0, x_norm.min()
        x_ = self.ap.denormalize(x_norm)
        assert (x - x_).sum() < 1e-3

        self.ap.signal_norm = True
        self.ap.symmetric_norm = True
        self.ap.max_norm = 1.0
        x_norm = self.ap.normalize(x)
        print(
            f" > MaxNorm: {self.ap.max_norm}, ClipNorm:{self.ap.clip_norm}, SymmetricNorm:{self.ap.symmetric_norm}, SignalNorm:{self.ap.signal_norm} Range-> {x_norm.max()} --  {x_norm.min()}"
        )

        assert (x_old - x).sum() == 0
        assert x_norm.max() <= self.ap.max_norm, x_norm.max()
        assert x_norm.min() >= -self.ap.max_norm, x_norm.min()  # pylint: disable=invalid-unary-operand-type
        assert x_norm.min() < 0, x_norm.min()
        x_ = self.ap.denormalize(x_norm)
        assert (x - x_).sum() < 1e-3

    def test_scaler(self):
        scaler_stats_path = os.path.join(get_tests_input_path(), "scale_stats.npy")
        conf.stats_path = scaler_stats_path
        conf.preemphasis = 0.0
        conf.do_trim_silence = True
        conf.signal_norm = True

        ap = AudioProcessor(**conf)
        mel_mean, mel_std, linear_mean, linear_std, _ = ap.load_stats(scaler_stats_path)
        ap.setup_scaler(mel_mean, mel_std, linear_mean, linear_std)

        self.ap.signal_norm = False
        self.ap.preemphasis = 0.0

        # test scaler forward and backward transforms
        wav = self.ap.load_wav(WAV_FILE)
        mel_reference = self.ap.melspectrogram(wav)
        mel_norm = ap.melspectrogram(wav)
        mel_denorm = ap.denormalize(mel_norm)
        assert abs(mel_reference - mel_denorm).max() < 1e-4

    def test_compute_f0(self):  # pylint: disable=no-self-use
        ap = AudioProcessor(**conf)
        wav = ap.load_wav(WAV_FILE)
        pitch = ap.compute_f0(wav)
        mel = ap.melspectrogram(wav)
        assert pitch.shape[0] == mel.shape[1]